https://jira.opendaylight.org This file is an XML representation of an issue en-us 8.20.10 820010 22-06-2022 https://jira.opendaylight.org/browse/OVSDB-156 ovsdb <p>If I have an OVS instance with "set-manager" set to point to the controller <br/> ovs-vsctl set-manager tcp:&lt;ip&gt;:6640</p> <p>THEN I remove journal etc (persisted data) and start controller I see following over and over again:<br/> <a href="http://pastebin.com/TcwBS9dr" class="external-link" target="_blank" rel="nofollow noopener">http://pastebin.com/TcwBS9dr</a></p> <p>If I reverse that order:<br/> (start ODL with cleaned out journal etc, wait until initialized)<br/> THEN<br/> set-manager</p> <p>I do not.</p> <p>What I see instead is over and over:<br/> 2015-05-17 09:32:31,512 | WARN | ds-oper-thread-0 | TransactionProxy | 179 - org.opendaylight.controller.sal-distributed-datastore - 1.2.0.SNAPSHOT | Failed to acquire operation permit for transaction member-1-txn-85</p> <p>Every 5 seconds.</p> <p>Note I haven't done anything at all except load my features in either case. There is no config other than a bridge in either case, and NO PERSISTED STATE in any case.</p> <p> </p> <p>Operating System: All<br/> Platform: All</p> OVSDB-156

Clustering: Errors if controller started after OVS configured with set-manager

Bug Resolved Done Unassigned Keith Burns Sun, 17 May 2015 16:34:03 +0000 Mon, 14 Jan 2019 08:32:37 +0000 Fri, 12 Jun 2015 00:58:54 +0000 unspecified Southbound.Open_vSwitch 0 10 <p>I think that this might be fixed by this:</p> <p><a href="https://git.opendaylight.org/gerrit/#/c/20619/" class="external-link" target="_blank" rel="nofollow noopener">https://git.opendaylight.org/gerrit/#/c/20619/</a></p> <p>Basically, you are telling karaf to start sfc-ovs <b>before</b> ovsdb-southbound.</p> <p>This patch tells karaf to start ovsdb-southbound <b>before</b> sfc-ovs.</p> <p>I can't quite explain <b>everything</b> I see in the logs from this, but it has a pretty good chance of fixing the issue. Would someone please test it?</p> <p>Patch for the "Failed to acquire operation permit" issue</p> <p><a href="https://git.opendaylight.org/gerrit/#/c/20680/" class="external-link" target="_blank" rel="nofollow noopener">https://git.opendaylight.org/gerrit/#/c/20680/</a></p> <p>Seems as duplicate of <a href="https://jira.opendaylight.org/browse/OPNFLWPLUG-442" title="redesign ofp inventory looses switches after few thousands of flows configured; often a problem to connect 25 switches" class="issue-link" data-issue-key="OPNFLWPLUG-442"><del>OPNFLWPLUG-442</del></a> or as Moiz pointed out related to <a href="https://jira.opendaylight.org/browse/OPNFLWPLUG-442" title="redesign ofp inventory looses switches after few thousands of flows configured; often a problem to connect 25 switches" class="issue-link" data-issue-key="OPNFLWPLUG-442"><del>OPNFLWPLUG-442</del></a>.</p> <p>I still have the same issue.</p> <p>Note we have EVERY SFC feature loaded and Ed had a patch <a href="https://git.opendaylight.org/gerrit/#/c/20619/" class="external-link" target="_blank" rel="nofollow noopener">https://git.opendaylight.org/gerrit/#/c/20619/</a> to address it. When I tested this on Tuesday <img class="emoticon" src="https://jira.opendaylight.org/images/icons/emoticons/help_16.png" height="16" width="16" align="absmiddle" alt="" border="0"/> the patch didn't resolve the issue.</p> <p>They key here is if there is an OVS switch configured and has set-manager configured, this happens every time you bring the controller up..</p> <p>This is a big problem for us with DevStack.</p> <p>Known facts:</p> <ul class="alternate" type="square"> <li>OVSDB stores remote-ids on OVS instances (instance-identifier to datastore)</li> <li>Journal is deleted-</li> </ul> <p>Theory:</p> <ul class="alternate" type="square"> <li>OVS reconnects to controller after journal is deleted, OVSDB learns state from<br/> Device along with remote-id there from previous session - this remote ID contains<br/> instance-identifier to datastore. OVSDB southbound tries to update data store,<br/> but does not expect it wiped (which happened when journal was created).</li> </ul> <p>So this could be OVSDB southbound bug - when remote-id is present before anything was configured to controller.</p> <p>@eaw: could it be this?</p> <p>Tony, you maybe onto something:</p> <p>When I do:<br/> sudo ovs-vsctl del-manager<br/> sudo ovs-vsctl set-manager tcp:192.168.50.1:6640</p> <p>I get endless:<br/> <a href="http://pastebin.com/xe9EiXnW" class="external-link" target="_blank" rel="nofollow noopener">http://pastebin.com/xe9EiXnW</a></p> <p>Seems this is ovsdb related somehow rather than clustering or md-sal.<br/> Moving it back to ovsdb so ovsdb folks are aware of that.</p> <p>I believe this is related to <a href="https://jira.opendaylight.org/browse/OVSDB-157" title="Use of emphemeral ports in iids causes breakage on reconnect" class="issue-link" data-issue-key="OVSDB-157"><del>OVSDB-157</del></a> and <a href="https://jira.opendaylight.org/browse/OVSDB-158" title="Config iids do not match oper iids for ovsdb-node, ovsdb-bridge, and termination-point" class="issue-link" data-issue-key="OVSDB-158"><del>OVSDB-158</del></a></p> <p>Should be fixed by<br/> <a href="https://git.opendaylight.org/gerrit/#/c/21484/" class="external-link" target="_blank" rel="nofollow noopener">https://git.opendaylight.org/gerrit/#/c/21484/</a></p> Blocks OPNFLWPLUG-442 Development External issue ID 3238 External issue URL Priority Rank 0|i020zr: